Abstract 3221: T-cell receptor sequencing for pharmacodynamic and response biomarkers of checkpoint blockade

Abstract

Background: Immunomodulatory cancer drugs such as anti-programmed death 1 (PD-1) and anti-cytotoxic T-lymphocyte antigen 4 (CTLA4) antibodies have shown significant clinical benefits across multiple indications. However, not all patients benefit from checkpoint blockade (CB) treatment strategies, highlighting the need for predictive biomarkers of response. Features of the T-cell receptor (TCR) repertoire have been reported to correlate with CB treatment and response (1). Here, we performed a comprehensive, retrospective analysis of TCR repertoires from different tissues (blood and tumor) at different time points (baseline and on treatment). To the best of our knowledge, our study, comprising a total of 558 samples across 3 tumor types (non-small cell lung cancer [NSCLC], renal cell carcinoma [RCC], and melanoma), is the largest analysis of TCR-sequencing data in CB clinical trials to date. Methods: We performed TCR sequencing on formalin-fixed paraffin-embedded tissue samples or peripheral blood samples using immunoSEQ Assays (Adaptive Biotechnologies Corp.) to profile the immune repertoire of patients from 3 clinical trials of nivolumab (anti-PD-1) or nivolumab + ipilimumab (anti-CTLA4) (NCT02041533, NCT01358721, NCT01621490). Repertoire clonality and clonal expansion statistics were computed, compared, and tested for their association with clinical activity. Results: We characterized the TCR repertoire from peripheral blood at baseline for 263 patients (209 NSCLC, 54 RCC) with matched on-treatment profiles for 49 patients (49 RCC). In addition, we profiled tumor TCR repertoires at baseline and on treatment for 127 patients (54 RCC, 73 melanoma) at time points predefined in each study’s protocol. Our analyses show that clonal expansion was accompanied by a comparable contraction of other clones and thus was not directionally enriched upon CB treatment. Moreover, comparison of baseline and on-treatment repertoire clonality did not reveal a significant shift following CB treatment. Results were consistent in blood and tumor tissue and across response groups. Baseline peripheral blood TCR clonality was not associated with best overall response or progression-free survival. Conclusions: Our results highlight that TCR repertoires are highly dynamic and that the currently established metrics are difficult to interpret or implement clinically as pharmacodynamic or response biomarkers. The variability in TCR clonality modulation upon CB treatment underscores the challenge of selecting the adequate on-treatment time point. Thus, our comprehensive analysis highlights the need for a better understanding of the TCR repertoire dynamics and development of methods to identify and functionally annotate tumor-specific TCRs. Reference: 1. Riaz N et al. Cell 2017;171:934-49 Citation Format: Simon Papillon-Cavanagh, Alice M. Walsh, Zhenhao Qi, Megan Wind-Rotolo, Abdel Saci, Parul Doshi, Radu Dobrin, Joseph Szustakowski. T-cell receptor sequencing for pharmacodynamic and response biomarkers of checkpoint blockade [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 3221.